IAP Family
The family of inhibitor of apoptosis (IAP) proteins comprises 8 members, XIAP, cIAP1, cIAP2, NAIP, ILP2, ML-IAP, survivin and BRUCE (also known as apollon). Members of the IAP family have been shown to inhibit programmed cell death through their ability to directly inhibit members of the caspase family of apoptotic enzymes, although the precise role of all 8 members is yet to be fully defined. The common structural feature of all IAP family members is a ˜70 amino acid zinc-binding fold termed the baculovirus IAP repeat (BIR) domain, which is present in one to three copies.
Many interactions between IAPs and other proteins are mediated via a surface groove on the BIR domain. BIR domains may be classified by their peptide-binding specificity. There are three types of BIR domains; type III domains (capable of binding caspase (and caspase-like) peptides with a specificity for proline in the third (P3) position (e.g. XIAP BIR3), type II domains (like type III domains but lacking the proline requirement e.g. XIAP BIR2) and type I domains (which do not bind caspases or similar peptides, e.g. XIAP BIR1) (Eckelman et al. Cell Death and Differentiation 2008; 15: 920-928). BIRs are small (˜70 amino acids) Zn-coordinated domains and a variety of proteins use their N-terminal to interact with the BIR domains grooves. BIR antagonists prevent caspases binding to BIRs and hence result in increased caspase activity thereby inducing auto-ubiquitination and proteasomal degradation of IAPs.
IAPs are overexpressed in many cancers including renal, melanoma, colon, lung, breast, ovarian and prostate cancers (Tamm et al., Clin. Cancer Research 2000; 6(5): 1796-803), and have been implicated in tumour growth, pathogenesis and resistance to chemo- and radio-therapy (Tamm 2000).
XIAP
XIAP is a 57 kDa protein with three BIR domains, the second and third of which bind caspases and a RING-type zinc finger (E3 ligase). XIAP binds several proteins in addition to caspases, including ligation substrates such as TAK1 and cofactor TAB1, MURR1 involved in copper homeostasis (Burstein et al., EMBO 2004; 23: 244-254), endogenous inhibitors such as second mitochondria-derived activator of caspases (SMAC), and those of less clear function such as MAGE-D1, NRAGE (Jordan et al., J. Biol. Chem. 2001; 276: 39985-39989).
The BIR3 domain binds and inhibits caspase-9, an apical caspase in the mitochondrial pathway of caspase activation. A groove on the surface of the BIR3 domain interacts with the N-terminus of the small subunit of caspase-9, locking capsase-9 in its inactive monomeric form with an incompetent catalytic site (Shiozaki et al., Mol. Cell 2003; 11: 519-527).
In addition to caspase-binding, XIAP also inhibits apoptosis through other mechanisms. XIAP forms a complex with TAK1 kinase and its cofactor TAB1 that leads to activation of JNK and MAPK signal transduction pathways, in turn leading to activation of NF-κB (Sanna et al., Mol Cell Biol 2002; 22: 1754-1766). XIAP also activates NF-κB by promoting NF-κB translocation to the nucleus and degradation of IκB (Hofer-Warbinek et al., J. Biol. Chem. 2000; 275: 22064-22068, Levkau et al., Circ. Res. 2001; 88: 282-290).
Cells transfected with XIAP are able to block programmed cell death in response to a variety of apoptotic stimuli (Duckett et al., EMBO 1996; 15: 2685-2694, Duckett et al., MCB 1998; 18: 608-615, Bratton, Lewis, Butterworth, Duckett and Cohen, Cell Death and Differentiation 2002; 9: 881-892).
XIAP is ubiquitously expressed in all normal tissues, but it is pathologically elevated in many acute and chronic leukaemias, prostate, lung, renal, and other types of tumours (Byrd et al., 2002; Ferreira et al., 2001; Hofmann et al., 2002; Krajewska et al., 2003; Schimmer et al., 2003; Tamm et al., 2000). In de novo acute myeloid leukaemia (AML), XIAP expression correlates with myelomonocytic French-American-British (FAB) subtypes M4/M5 (P<0.05) and expression of monocytic markers in AML blasts. In addition, XIAP was found to be overexpressed in normal monocytes but undetectable in granulocytes. In AML, XIAP expression was significantly lower in patients with favourable rather than intermediate or poor cytogenetics (n=74; P<0.05) (Tamm et al., Hematol. J. 2004; 5(6): 489-95).
Overexpression renders cells resistant to multi-agent therapy and is associated with poor clinical outcome in disease including AML, renal cancer, melanoma (Tamm et al., Clin. Cancer Research 2000; 6: 1796-1803) and lung cancer (Hofmann et al., J. Cancer Res. Clin. Oncology 2002; 128(10): 554-60).
XIAP is translated by a cap-independent mechanism of translation initiation that is mediated by a unique internal ribosome entry site (IRES) sequence element located in its 5′ untranslated region. This allows XIAP mRNA to be actively translated during conditions of cellular stress when the majority of cellular protein synthesis is inhibited. Translational upregulation of XIAP in response to stress increases resistance to radiation induced cell death (Holcik et al., Oncogene 2000; 19: 4174-4177).
XIAP inhibition has been investigated in vitro via several techniques including RNA silencing, gene knockout, peptidic ligand mimetics and small molecule antagonists, and has been shown to promote apoptosis as a monotherapy and to sensitise many tumour types to chemotherapy, including bladder (Kunze et al., 2008; 28(4B): 2259-63). XIAP knockout mice are born at the expected Mendelian frequency, with no obvious physical or histological defects, and normal life spans (Harlin et al., Mol. Cell Biol. 2001; 21(10): 3604-3608). This indicates that lacking XIAP activity is not toxic in normal tissues and suggests a therapeutic window over tumour cells. Further studies have shown XIAP is a critical discriminator between apoptosis in type 1 and type 2 cells including hepatocytes and therefore should be used with caution in patients with underlying liver conditions (Jost et al., Nature, 2009, 460, 1035-1041). It was noted that the cIAP1 and cIAP2 levels are upregulated in the XIAP knockout mouse and may protect from pathology via a compensatory mechanism, suggesting pan-inhibition may be required for functional knockout. Similarly, cIAP1 and cIAP2 knockout mice are also asympotomatic (Conze et al., Mol. Biol. Cell 2005; 25(8): 3348-56). While lack of any one of the IAPs produced no overt phenotype in mice, deletion of cIAP1 with cIAP2 or XIAP resulted in mid embryonic lethality (Moulin, EMBO J., 2012).
Endogenous IAP antagonists such as SMAC have been used to validate members of this family as targets for therapeutic agents. SMAC peptides chemosensitise tumour cells, and in combination with platins and Tumour Necrosis Factor α-related apoptosis inducing ligand (TRAIL) in xenografts, results in tumour growth delay (Fulda et al., Nat. Med. 2002; 808-815; Yang et al., Cancer Res. 2003; 63: 831-837).
A natural product, embellin, was identified as binding at the surface groove of the BIR3 domain of XIAP with similar affinity to the natural SMAC peptide. Embellin induces apoptosis in cell lines in vitro and results in tumour growth delay in xenografts (Nikolovska-Coleska et al., J. Med. Chem. 2004; 47(10): 2430-2440; Chitra et al., Chemotherapy 1994; 40: 109-113).
XIAP antisense oligonucleotides have been developed as therapeutic agents for solid tumour and haematological malignancies. In vitro these antisense oligonucleotides have been shown to knockdown protein expression levels by ˜70%, induce apoptosis and sensitise cells to chemotherapy and delay tumour growth in vivo. One of these agents, AEG351156, has been studied in clinical trials (Hu et al., Clin. Cancer Res. 2003; 9: 2826-2836; Cummings et al., Br. J. Cancer 2005; 92: 532-538).
Small molecule antagonists of XIAP developed include peptidomimetics as well as synthetic agents. The peptidomimetics target the BIR3 domain, mimicking SMAC disruption of caspase-9 binding to XIAP, have shown induction of apoptosis in a variety of tumour cell lines as a single agent, as well as chemosensitisers and are being further investigated clinically (Oost et al., J. Med. Chem. 2004; 47: 4417-4426; Sun et al., Bioorg. Med. Chem. Lett. 2005; 15: 793-797).
Synthetic small molecule antagonists of BIR3 and BIR2 domains also demonstrate anti-tumour activity in several different models, including induction of apoptosis by annexin-V staining and IC50s of <10 μM against over one-third of the NCl60 cell line panel. XIAP antagonists also induced dose-dependent cell death of primary-cultured leukaemia cells in 5 out of 5 chronic lymphocytic leukaemia cell lines and 4 out of 5 acute myeloid leukaemia cell lines (Schimmer et al., Cancer Cell 2004; 5: 25-35; Berezovskaya et al., Cancer Res. 2005; 65(6): 2378-86).
High levels of XIAP protein in tumour cell lines were inversely correlated with sensitivity to some anti-cancer drugs, particularly cytarabine and other nucleosides (Tamm et al., Clin. Cancer Research 2000; 6: 1796-1803). XIAP inhibition potentiates TRAIL-induced antitumor activity in two preclinical models of pancreatic cancer in vivo (Vogler 2008). Gene expression and transfection studies suggest that the increased expression of apoptosis suppressor XIAP plays an important role in anoikis resistance and in the survival of circulating human prostate carcinoma cells, thereby promoting metastasis. Small molecule antagonists were found to be anti-metastatic in these models (Berezovskaya et al., Cancer Res. 2005; 65(6): 2378-86).
XIAP has also been found to be involved in other pathways associated with cancer and other diseases and these may also benefit from XIAP targeted agents. The E3 ligase activity of the RING finger domain of XIAP is able to bind both to TAB1 and to an upstream BMP receptor (type 1), suggesting that XIAP may signal in a TGF-β-mediated pathway (Yamaguchi et al., EMBO 1999; 179-187). Focal adhesion kinase (FAK) overexpression has been shown to result in upregulated XIAP expression (Sonoda et al., J. Biol. Chem. 2000; 275: 16309-16315). E3 ligases are attractive therapeutic targets and molecules which target this activity in other proteins such as MDM2 are being developed (Vassilev et al., Science 2004; 303: 844-848). Direct or indirect inhibition of the XIAP ligase activity may also be useful in the treatment of cancer and other diseases. Dysregulated apoptotic signalling, which would result from inhibition of IAP function in controlling programmed cell death, has also been implicated in many diseases, including disorders associated with cell accumulation (e.g. cancer, autoimmunity, inflammation and restenosis) or disorders where excessive apoptosis results in cell loss (e.g. stroke, heart failure, neurodegeneration such as Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis, AIDS, ischaemia (stroke, myocardial infarction) and osteoporosis).
XIAP is an important apoptotic regulator in experimental autoimmune encephalomyelitis and a potential pharmacological target for treating autoimmune diseases such as multiple sclerosis (MS) (Moore et al., 2004; 203(1): 79-93). Antisense-mediated knockdown of XIAP reverses paralysis in an animal model of MS suggesting that treatments targeting XIAP, and perhaps other IAPs, may have utility in the treatment of MS (Hebb et al., Curr. Drug Disc. Tech. 2008; 5(1): 75-7).
cIAP1, cIAP-2, XIAP and survivin are overexpressed in malignant pleural mesothelioma and are responsible for a large degree of the resistance of cultured mesothelioma cells to cisplatin. Levels of circulating TNF-α are significantly higher in mesothelioma patients prior to surgical tumor debulking compared with those after surgery. TNF-α increases mRNA and protein levels of IAP-1, IAP-2 and XIAP (Gordon et al., 2007). NF-κB upregulation plays an important survival role in mesotheliomas in response to the inflammatory effects of exposure to asbestos fibres (Sartore-Bianchi et al., 2007). IAP antagonists have the potential to reverse the pro-survival effect of TNF-α.
The ability of cell lines to upregulate TNF-alpha expression sufficiently to act in an autocrine fashion and kill the cells, once cIAP1 & 2 are depleted, is believed to be important for IAP activity (Nature Reviews Cancer (2010), 10(8), 561-74, Gryd-Hansen, M). In vivo, however, certain tumour types are surrounded by a pro-inflammatory cytokine network and hence the tumour cells which, on depletion of cIAP1/2 are switched towards cell killing by apoptosis, may be triggered to apoptose by TNF-alpha (or other Death Receptor cytokine agonists) already being produced by surrounding cells in the tumour microenvironment, such as tumour-associated macrophages, or indeed by the tumour cells themselves. Certain tumour types such as breast, ovarian and melanoma display this “inflammatory phenotype” which could potentially be targeted by IAP antagonists.
cIAP1 and cIAP2
Cellular IAP (cIAP) 1 and 2 are closely related members of the IAP family with three BIR domains, a RING domain and a caspase-recruitment (CARD) domain. A functional nuclear export signal exists within the CARD domain of cIAP1 which appears to be important for cell differentiation (Plenchette et al., Blood 2004; 104: 2035-2043). The presence of this CARD domain is unique to cIAP1 and cIAP2 within the IAP family of proteins. These two genes reside in tandem on chromosome 11q22 and given their high degree of similarity are thought to have arisen via gene duplication.
cIAP1, like XIAP and survivin, is widely expressed in tumour cell lines, and has been found to be expressed at high levels in colorectal cancers in particular, as well as lung, ovarian, renal, CNS and breast cancers (Tamm et al., Clin. Cancer Res. 2000; 6: 1796-1803). cIAP2 expression is generally more restricted and is thought to be regulated though constitutive ubiquitination and degradation by cIAP1 (Conze et al., Mol. Biol. Cell 2005; 25(8): 3348-56; Mahoney et al., PNAS 2008; 105: 11778-11783). Immunohistochemistry and western blot analysis identified cIAP1 and cIAP2 as potential oncogenes as both are overexpressed in multiple lung cancers with or without higher copy numbers (Dia et al., Human Mol. Genetics 2003; 12(7): 791-801). cIAP1 expression level preferentially seems to play an important role in low-stage adenocarcinoma (Hofmann et al., J. Cancer Res. Clin. Oncology 2002; 128(10): 554-60).
Increased levels of cIAP1 and cIAP2 and reduced levels of endogenous inhibitors are associated with chemoresistance as has been seen for XIAP. cIAP overexpression has been found to correlate in vitro to resistance to DNA alkylating agents such as carboplatin, cisplatin and topoisomerase inhibitor VP-16 (Tamm et al., Clin. Cancer Res. 2000; 6: 1796-1803). Levels of cIAP1 and survivin were found to be high in thyroid cancer cells after cisplatin and doxorubicin treatment. Cells resistant to chemotherapy such as taxol showed reduced expression of SMAC and released minimal amounts of this protein from the mitochondria. Down-regulation of cIAP1 and survivin has been found to increase the cytotoxicity of cisplatin and doxorubicin, whereas overexpression of SMAC improved the efficacy of taxol. However, silencing of cIAP1 and survivin by RNA interference restored sensitivity to doxorubicin and cisplatin (Tirrò et al.; Cancer Res. 2006; 66(8): 4263-72).
SMAC mimetics such as LBW242 were originally thought to primarily target XIAP. However studies have shown that cIAP1 was targeted for degradation by autoubiquitination in cells (Yang et al., J. Biol. Chem. 2004; 279(17): 16963-16970) and may have contributed to the apoptotic effects that resulted. SiRNA of cIAP1 and Tumour Necrosis Factor (TNF)-alpha induction (or stimulation) were found to combine synergistically and render cell lines more sensitive (Gaither et al. Cancer Res. 2007; 67 (24): 11493-11498).
cIAP1 and cIAP2 have been demonstrated to be critical regulators of the NF-κB signalling pathway which is involved in a diverse range of biological processes, particularly in innate and adaptive immunity as well as in proliferation and survival. NF-κB pathway deregulation is associated with inflammation and cancers including hepatitis and ulcerative colitis, gastritis, hepatocellular carcinoma colorectal cancer and gastric cancers, as well as angiogenesis and metastasis (Shen et al., Apoptosis 2009; 14: 348-363).
On ligand binding, the TNF Receptor (TNF-R) recruits TNFR-associated Death Domain (TRADD) and receptor-interacting protein (RIP) 1. TRAF2 and cIAP1/cIAP2 are then recruited to form a large membrane complex. RIP1 is ubiquitinated and these polyubiquitin chains serve as a docking site for downstream kinases, resulting in NF-κB pathway signalling effects (Ea et al., Mol. Cell 2006; 22: 245-257; Wu et al., Nat. Cell Biol. 2006; 8: 398-406). The extended roles are complex and yet to be fully defined but cIAP1 and cIAP2 are identified as key components of TNF-alpha mediated NF-κB signalling regulation as well as constitutive (ligand-independent/classical) NF-κB signalling (Varfolomeev et al., Cell 2007; 131(4): 669-81). cIAP1 and cIAP2 have been shown to bind TRAF2, an adapter protein that functions in both the classical and alternative NF-κB pathways as well as MAPK pathway signalling pathway (Rothe et al., Cell 2005; 83: 1243-1252). cIAP1 and cIAP2 directly target RIP1 for ubiquitination in vitro (Betrand et al., Mol. Cell 2008; 30: 689-700).
TNF-alpha regulates many cellular functions, including apoptosis, inflammation, immune response, and cell growth and differentiation (Trace et al., Annu. Rev. Med. 1994; 45: 491-503) and therapeutic IAP antagonists may be of benefit in conditions where these functions are affected.
Production of TNF-alpha is seen in many malignant tumours, and is one of the key drivers of cancer-related inflammation that drives tumour development and/or progression. cIAPs protect cancer cells from the lethal effects of TNF-alpha.
NAIP
NAIP was the first IAP to be discovered (Roy et al., Cell 1995; 80: 167-178). NAIP is unique among the IAPs in that it possesses a nucleotide-binding and oligomerisation domain, as well as leucine rich repeats which are similar to those contained in proteins normally involved in innate immunity. There are indications that NAIP may also be over expressed in some cancers including breast and oesophageal cancer (Nemoto et al., Exp. Mol. Pathol. 2004; 76(3): 253-9) as well as MS (Choi et al., J. Korean Med. 2007; 22 Suppl: S17-23; Hebb et al., Mult. Sclerosis 2008; 14(5): 577-94).
ML-IAP
Melanoma inhibitor of apoptosis protein (ML-IAP) contains a single BIR and RING finger motif. ML-IAP is a powerful inhibitor of apoptosis induced by death receptors and chemotherapeutic agents, probably functioning as a direct inhibitor of downstream effector caspases (Vucic et al., Curr. Biol. 2000; 10(21): 1359-66). ML-IAP is also known as Baculoviral IAP repeat-containing protein 7 (BIRC7), Kidney inhibitor of apoptosis protein (KIAP), RING finger protein 50 (RNF50) and Livin. The BIR domain of ML-IAP possesses an evolutionarily conserved fold that is necessary for anti-apoptotic activity. It has been found that the majority of melanoma cell lines express high levels of ML-IAP in contrast to primary melanocytes, which expressed undetectable levels. These melanoma cells were significantly more resistant to drug-induced apoptosis. Elevated expression of ML-IAP renders melanoma cells resistant to apoptotic stimuli and thereby potentially contributes to the pathogenesis of this malignancy.
ILP-2
ILP-2, also known as BIRC8, has a single BIR domain and a RING domain. ILP-2 is expressed only in testis in normal cells, and binds to caspase 9 (Richter et al, Mol. Cell. Biol. 2001; 21: 4292-301).
Survivin
Survivin, also known as BIRC5, inhibits both caspase 3 and caspase 7, but its primary function is mitotic progression regulation, rather than the regulation of apoptosis. Survivin promotes formation of microtubules in the mitotic spindle, counteracting apoptosis during cell cycle. Apoptosis inhibition by survivin is predictive of poor outcome in colorectal cancer (Kawasaki et al., Cancer Res. 1998; 58(22): 5071-5074) and stage III gastric cancer (Song et al., Japanese J. Clin. Oncol. 2009; 39(5): 290-296).
BRUCE
BRUCE (BIR repeat-containing ubiquitin-conjugating enzyme) is a peripheral membrane protein in the trans-Golgi network with a single BIR domain, most similar to that of survivin. BRUCE is inhibited via three mechanisms: (i) SMAC binding, (ii) HtrA2 protease and (iii) caspase-mediated cleavage. In addition, BRUCE acts as a E2/E3 ubiquitin ligase via ubiquitin-conjugating (UBC) domain.